Probabilistic discrimination between large-scale environments of intensifying and decaying African Easterly Waves

Despite the fact that about nearly 85% of intense hurricanes have their origins as African Easterly Waves (AEWs), determining a priori the likelihood of AEW intensification remains a difficult task. In this study we have developed a Bayesian methodology to understand genesis of North Atlantic TCs spawned by AEWs through the examination of the characteristics of the AEW itself together with the large-scale environment, resulting in a probabilistic discrimination between large-scale environments associated with intensifying and decaying AEWs. The methodology is based on an objective AEW tracking scheme used for the period 1980 to 2001 based on spatio-temporally Fourier-filtered relative vorticity and meridional winds at different levels and outgoing long wave radiation.

Using the AEW and HURDAT data sets, probability density functions (PDFs) of environmental variables that discriminate between AEWs that decay, become TCs or become major hurricanes are determined. Results indicate that the initial amplitude of the AEWs is a major determinant for TC genesis, and that TC genesis risk increases when the wave enters an environment characterized by pre-existing large-scale convergence and moist convection. For the prediction of genesis, the most useful variables are column integrated heating, vertical velocity and specific humidity, and a combined form of divergence and vertical velocity and SST.

The Bayesian methodology is also used to explore the role of aerosol loading in AEW intensification as well as to compare easterly wave-associated hurricane genesis in the Atlantic versus the Eastern Pacific.